> ## Documentation Index
> Fetch the complete documentation index at: https://docs.citino.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# The ± band and n

> What the two small marks beside every number mean, and when a number is still early.

Every number Citino measures from answers comes with two marks: `n`, how many answers it rests on, and `±`, how far it could be off. Read them before you read the number.

## n

> How many answers a number is based on. The more answers, the more you can trust it.

`n` counts answers, not questions. Each question is asked on every engine you track every day, so `n` grows every day.

Some numbers only count some answers:

* **Visibility and share of voice** count every answer in scope.
* **Position** counts only the answers that named you.
* **Sentiment** counts only the answers that named you and described you one way or the other.

## ±

> How far the true number could reasonably be from the one shown. It is wide when there are few answers or few questions behind it, and narrows as more come in.

We work the range out at 95%: wide enough that the true figure is very likely inside it.

### Why it depends on questions, not only answers

Answers to the same question tend to agree with each other. Ask one question on many days and you get much the same brands each time. So a pile of answers to a few questions tells you less than the count suggests.

That is why we work out the range by resampling whole questions rather than single answers. It gives a wider range than treating every answer as independent, and an honest one. Two things follow:

* **One question on its own has no range.** As the app puts it: "A single question has no ± range; ranges are worked out across many questions." Where there is no range to show, you see `±—`.
* **A project with few questions keeps a wider range** than one with many.

The calculation is fixed, so the same answers always give the same `±`. A range only changes when the answers behind it do. [How bands are computed](https://docs.citino.ai/guides/methodology/bands.md) has the method.

## Low sample

> This number rests on only a few answers, so read it as an early signal rather than a settled result. It firms up as we collect more each day.

The badge appears in three places:

* on **Prompts**, beside one question's visibility, below 15 answers;
* in **Overview**'s Segments table, beside a segment's visibility, below 30 answers;
* on **Objections**, beside an objection's score, below 30 answers.

Position and sentiment carry no badge, so read their `n` yourself.

**A low sample never hides a number.** From the first answer, the figure is shown with its `±` and `n`, and the badge sits beside it as a quiet cue.

## When there is no number

| You see | It means |
| - | - |
| `—` with "needs one full run" | No answers in scope yet. Nothing has been measured. |
| `0% · not named` | There are answers and none named you. A finding, not a gap in the data. |
| `—` in an unclaimed segment | The segment is outside your scope, so it is shown for context and not scored. See [Scope and segments](https://docs.citino.ai/guides/scope-and-segments.md). |

## Related

> [card] [Within noise](https://docs.citino.ai/guides/within-noise.md)
>
> Why a change smaller than its ± is not a move.

> [card] [How bands are computed](https://docs.citino.ai/guides/methodology/bands.md)
>
> The method, in more detail.

> [card] [Visibility](https://docs.citino.ai/guides/visibility.md)
>
> The headline number these marks sit beside.

> [card] [Glossary](https://docs.citino.ai/guides/glossary.md)
>
> Every term, in the app's words.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.